The Wisdom of Crowds and Selecting an Investment Manager

The Wisdom of CrowdsI just read James Surowiecki’s The Wisdom of Crowds.   The provocative title refers to the idea that, given the right circumstances, a group of people will consistently reach decisions which are better than the decisions that the smartest people in the group would be able to reach on their own. 

As someone who considers himself an investing expert, it might be surprising that I like this book.  After all, he frequently cites examples from investing as to how bad the “experts” are.  For instance, a vast majority of mutual fund managers under-perform their benchmarks in any given year, as well as over longer periods. 

Surowiecki’s conclusion is that the crowd (in this case, the market as a whole) is smarter than most mutual fund managers, and even if there are managers who are smarter than the market (The namesake of my cat, Warren Buffett, being a prime example), the amount of effort necessary to distinguish a superior manager (if it is possible to do at all, since past performance is not a guarantee of future results) is far in excess of the possible reward.

Why are most money mangers so bad at the job they are paid to do?  Because it is more risky for them to take non-consensus positions, than it is for them to buy the same stocks everyone else is buying, even if they know those stocks to be overvalued.

Surowiecki concludes that, while there may be superior investment managers who can consistently outperform the market, the effort necessary to identify those managers is likely to be prohibitive.  After all, past performance is no guarantee of future results, and we are as likely to be fooled by randomness into choosing a manager who is lucky as we are to choose a manager who is killed.

I disagree.  I feel it is possible to identify superior managers, and Surowiecki has laid out the tools to do so, even if he does not realize he has done so.  He shows quite convincingly that stock markets can be very efficient when the participants are diverse, and reach decisions independently.   As he also shows, most money managers fail on all counts.

Markets are efficient at finding the best prices for stocks when the participants are diverse and independent.  Since that is often not the case in the stock market, there will be opportunities for individuals to outperform the market.When looking for such an individual, what are the clues?

  • Any manager whose holdings mimic the entire market (i.e. are extremely diversified) can be ruled out.  His returns will also mimic market returns, and will likely even under perform, due to fees and commissions.
  • His choices should not be random; they should be based on private information or research that is not available to (or is being ignored by) the entire market.
  • There should be some reason to believe that this private information or research gives him some true insight into the value of his investments: a portfolio of five companies chosen at random will not perform the same as the market as a whole, but there is no reason to believe it will outperform, either.

If you believe, as James Surowiecki does, that the necessary effort is prohibitive to find a manager like this, the best investment strategy would be to index, that is try to mimic the entire market’s return for the lowest possible fees. 

To me, the possible gains from above-average performance seem well worth the effort necessary to find the few superior managers. Over a period of 20 years, the difference between a 7% annual compounded return and an 8% annual compounded return is about 25%.  In other words, you would get the same effect by starting with $100,000 and earning 8% for 20 years, as you would by starting with $80,000 and earning 7% over the same period.

 The trick is not to give too much weight to past performance, but instead concentrate on the factors which might allow a manager to acheive superior performance in the future.  As an example, here is how I would go about selecting a mutual fund based on the above criteria.  The same approach would apply to a hedge fund or private account manager, except for the fact that in this case, the search for such a manager might truly be outside the ability of most individuals, since the relevant data is not readily available in a searchable form.

  1. Have a small number of positions (since these are less likely to mimic the market,) or changing emphasis on particular industries or sectors which vary significantly from market weighting.
  2. Have low turnover (private information is hard to come by, and given the effort involved, it is not likely to lead to rapidly changing strategies.)
  3. Have a methodology that emphasizes independent research, and is easy to explain.  Black box approaches based on computer models can work for a time, but the market has a tendency to make them go rapidly obsolete.  This is a large part of the reason past performance is such a bad predictor of future results.  “Fundamental” investment strategies tend to pass this test, so long as they are based on deeper research than plucking various ratios from a company’s financial statements.
  4. Look for funds that don’t fit well into any one “Style” or capitalization category.  Independent and diverse thinkers will be hard to categorize.
  5. Funds with less than $1 billion under management.  When assets under management get too high, it becomes difficult to act on what private information you have without greatly moving the market.
  6. Look for low “Beta” compared to the fund’s category. Beta is a measure of correlation with the market as a whole, and can be a proxy for diversity.
  7. Make sure that the fees you are paying are justified by the excess returns (in comparison with the risk involved) the strategy is likely to yield.

Using the Yahoo! Mutual fund screener, with the following criteria: manager tenure > 5 years; no load; expense ratio < 1%, assets < $1 billion, turnover < 30%, is a good place to start.  This yields a reasonable number of funds to sort through, and the choices can be further narrowed by ignoring the index funds, and by checking Beta under “Risk.”

Happy Hunting!

1 Comment

  1. […] If I were to follow this strategy, I might even be able to get that 10x return in less than 10 years, but the most likely result would be turning $100,000 into $10,000.  So I guess I’m like all the other investment professionals, and I don’t expect this experiment work, especially in the current climate of over-valuation.  But then, most investment professionals aren’t good at managing money, either. […]

RSS feed for comments on this post

Comments are closed.

%d bloggers like this: